Peer-Reviewed Publication
Sci Rep2026;16(1)March 22, 2026Journal Article
Validation of conformal prediction in cervical atypia classification.
Misgina Tsighe Hagos1,2, Antti Suutala3, Dmitrii Bychkov3, Hakan Kücükel3, Joar von Bahr3,4,5, Milda Poceviciute6,7, Johan Lundin3,5, Nina Linder3,4, Claes Lundström6,7,8
1Department of Science and Technology, Linköping University, Norrköping, Sweden. misgina.tsighe.hagos@liu.se.
2Center for Medical Imaging Science and Visualization, Linköping University, Linköping, Sweden. misgina.tsighe.hagos@liu.se.
3Institute for Molecular Medicine Finland - FIMM, University of Helsinki, Helsinki, Finland.
4Department of Women's and Children's Health, Uppsala University, Uppsala, Sweden.
5Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden.
6Department of Science and Technology, Linköping University, Norrköping, Sweden.
7Center for Medical Imaging Science and Visualization, Linköping University, Linköping, Sweden.
8Sectra AB, Linköping, Sweden.
Abstract
Deep learning based cervical cancer classification can potentially increase access to screening in low-resource regions. However, deep learning models are often overconfident and do not reliably reflect diagnostic uncertainty. Moreover, they are typically optimized to generate maximum-likelihood predictions, which fail to convey uncertainty or ambiguity in their results. Such challenges can be add…
Create a free account to keep reading
Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.
Want unlimited research access? See Pro plans